본문 바로가기
  • Home

Analysis of Relative Clauses in English-to-Korean Machine Translation Systems’ Output

  • Modern English Education
  • Abbr : MEESO
  • 2024, 25(), pp.74-84
  • Publisher : The Modern English Education Society
  • Research Area : Humanities > English Language and Literature > English Language Teaching
  • Received : February 13, 2024
  • Accepted : March 27, 2024
  • Published : April 30, 2024

Kim, Chae-Eun 1

1부산대학교

Accredited

ABSTRACT

The use of artificial intelligence (AI)-based tools for teaching and learning English as a foreign language has become increasingly functional in Second Language Acquisition (SLA). This study aimed to explore the potential of English-to-Korean machine translators (MTs) as interactive tools for language learning, specifically for Korean EFL learners. The study assessed the accuracy of two commonly used MTs in converting English Relative Clauses (RCs) into Korean equivalents by analyzing translation data from research articles. A total of 60 articles provided 513 examples of RCs. The study analyzed the output of MTs in three ways: (i) whether the translated meaning accurately represented the targeted form, (ii) what types of errors the Korean translations made, and (iii) how the results compared to previous works on Korean-to-English translation of RCs. The results showed that currently available English-to-Korean MTs could produce semantically correct translations at high rates (around 95%). However, they make errors such as omitting lexical elements, resulting in structural ambiguity. Compared to previous works, the accuracy rate has gradually improved. These findings suggest that MTs have great potential as aids in English reading and writing instruction.

Citation status

* References for papers published after 2022 are currently being built.